← back to catalog · registered 2026-08-22 13:56

stephenlzc/dolphin-llama3-zh-cn-uncensored

stephenlzc Llama 8.0B GGUF 8K ctx
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/stephenlzc%2Fdolphin-llama3-zh-cn-uncensored"
Response includes
  • classification m-uncensored
  • files 18
  • hub_downloads_all_time 160,536
  • providers 1
  • author_summary 5 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
161K
6K last 30d - cooling
Likes
24
Model age
2.3y ago
created 2024-06-21
Available via
1 provider
featherless-ai

Training datasets

1 of 3 in /datasets

Corpora the author lists in the model card. Datasets tracked in our /datasets catalog carry a category badge linking to the workflow stage. Others open on Hugging Face.

Downloads over time
Now162.9K→from6.1K↑2,587%
059.7K119.4K179.1K6.1K on Jul 24, 2024162.9K on Oct 11Jul '24Nov '24Mar '25Jul '25Nov '25MarJul
Jul 24, 2024 → Oct 11 · 158 snapshots · spans 809 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Metadata

License
mit
Languages
zh en
Quantizations
BF16 F16 Q4_K Q8_0
Tags
transformers safetensors gguf llama text-generation text-generation-inference code unsloth NSFW conversational zh en

Related

Total size
57.6 GB
Files
18
Quantizations
5
Registered
2026-08-22 13:56
Last updated on HF
2024-06-24 17:44

Files by quantization

BF16 1 file 15.0 GB
dolphin-llama3-zh-cn-uncensored-unsloth.BF16.gguf 15.0 GB b830aae7 download
F16 1 file 15.0 GB
dolphin-llama3-zh-cn-uncensored-unsloth.F16.gguf 15.0 GB 6ab477aa download
Q8_0 1 file 7.95 GB
dolphin-llama3-zh-cn-uncensored-unsloth.Q8_0.gguf 7.95 GB c2cd77a7 download
Q4_K 1 file 4.58 GB
dolphin-llama3-zh-cn-uncensored-unsloth.Q4_K_M.gguf 4.58 GB 2c36ae9a download
Auxiliary files 14 files 15.1 GB
model-00002-of-00004.safetensors 4.66 GB 4e0a0c0e download
model-00001-of-00004.safetensors 4.63 GB b18c8da1 download
model-00003-of-00004.safetensors 4.58 GB 43d78dd7 download
model-00004-of-00004.safetensors 1.09 GB df88b85a download
adapter_model.safetensors 160 MB 002bf89a download
tokenizer.json 8.66 MB 7b4638ca download
tokenizer_config.json 50.1 KB 6def5af9 download
model.safetensors.index.json 23.4 KB 8d8b4acd download
.gitattributes 2.03 KB 902ace95 download
README.md 1.71 KB f1100243 download
config.json 768 B 05553e40 download
adapter_config.json 748 B bda85ed1 download
special_tokens_map.json 444 B 44e8cb89 download
generation_config.json 142 B 168486d1 download

README current version from Hugging Face


datasets:

  • Minami-su/toxic-sft-zh
  • llm-wizard/alpaca-gpt4-data-zh
  • stephenlzc/stf-alpaca
    language:
  • zh
  • en
    license: mit
    pipeline_tag: text-generation
    tags:
  • text-generation-inference
  • code
  • unsloth
  • NSFW
    task_categories:
  • conversational
    base_model: cognitivecomputations/dolphin-2.9-llama3-8b
    widget:
  • text: >-
    Is this review positive or negative? Review: Best cast iron skillet you will
    ever buy.
    example_title: Sentiment analysis
  • text: >-
    Barack Obama nominated Hilary Clinton as his secretary of state on Monday.
    He chose her because she had ...
    example_title: Coreference resolution
  • text: >-
    On a shelf, there are five books: a gray book, a red book, a purple book, a
    blue book, and a black book ...
    example_title: Logic puzzles
  • text: >-
    The two men running to become New York City's next mayor will face off in
    their first debate Wednesday night ...
    example_title: Reading comprehension

Model Details

Model Description

It's my first finetune exmaple.

Using cognitivecomputations/dolphin-2.9-llama3-8b as base model, and finetune with Minami-su/toxic-sft-zh and llm-wizard/alpaca-gpt4-data-zh to let tha model support Chinese.

Training Procedure

Open In Colab

Training Data

Base Model
🐬cognitivecomputations/dolphin-2.9-llama3-8b

Dataset

README history 14 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2024-06-24Update README.md2bf897b1.7 KB
    Loading...
  2. 2024-06-24Update README.mdeebfcce1.7 KB
    Loading...
  3. 2024-06-21Update README.mdf4292551.7 KB
    Loading...
  4. 2024-06-21Update README.mde3b58941.7 KB
    Loading...
  5. 2024-06-21Update README.md810253f1.7 KB
    Loading...
  6. 2024-06-21Update README.md77c3e6c1.7 KB
    Loading...
  7. 2024-06-21Update README.md6699d9d903 B
    Loading...
  8. 2024-06-21Update README.mdfbc43bf903 B
    Loading...
  9. 2024-06-21Upload model trained with Unslothd2c45ff283 B
    Loading...
  10. 2024-06-21Update README.md962a220272 B
    Loading...
  11. 2024-06-21Update README.mdb2d7065228 B
    Loading...
  12. 2024-06-21Update README.md2c7a381215 B
    Loading...
  13. 2024-06-21Update README.md35d56c665 B
    Loading...
  14. 2024-06-21initial commit8c9027024 B
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration